A method and respective system for determining quality of experience parameters of an encrypted video stream received at a client device is provided. The method comprises extracting, from one or more encrypted video streams sent over a network from a content server to a plurality of client devices, a first instance of at least one stream-related feature. A first instance of at least one quality-related label of a plurality of quality-related labels is determined based on applying a trained classifier to the first instance of the at least one stream-related feature, wherein each of the plurality of quality-related labels corresponds to a respective experience parameter of the quality of experience parameters of the encrypted video stream received at the client device.
Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
1. A method for determining quality of experience parameters of an encrypted video stream received at a client device, the method comprising: extracting, from one or more encrypted video streams sent over a network from a content server to a plurality of client devices, a first instance of at least one stream-related feature associated with at least one of throughput of the one or more encrypted video streams, peak density of the one or more encrypted video streams, or quantity of the one or more encrypted video streams; determining a first instance of at least one quality-related label of a plurality of quality-related labels based on applying a trained classifier to the first instance of the at least one stream-related feature extracted from the one or more encrypted video streams, the trained classifier being trained based on a training dataset, validated based on a validation dataset, and tested based on a test dataset, wherein each of the plurality of quality-related labels corresponds to a respective quality of experience parameter of the quality of experience parameters of the encrypted video stream received at the client device, and wherein the at least one quality-related label is associated with playback quality of the one or more encrypted video streams; changing one or more first network parameters associated with the network to perform capacity enhancements at a radio access network used to connect the content server to a first subset of the plurality of client devices via the network or at a network core used to connect the content server to the first subset of the plurality of client devices via the network; extracting, from the one or more encrypted video streams sent from the content server to the first subset of the plurality of client devices, a second instance of the at least one stream-related feature after the changing of the one or more first network parameters; determining a second instance of the at least one quality-related label based on applying the trained classifier to the second instance of the at least one stream-related feature extracted after the changing of the one or more first network parameters; analyzing the first instance of the at least one quality-related label determined before the changing of the one or more first network parameters and the second instance of the at least one quality-related label determined after the changing of the one or more first network parameters to determine a measurement of an impact of the changing of the one or more first network parameters on the quality of experience parameters of the encrypted video stream received at the first subset of the plurality of client devices; and changing one or more second network parameters associated with the network used to connect the content server to a second subset of the plurality of client devices in view of the measurement of the impact of the changing of the one or more first network parameters on the quality of experience parameters of the encrypted video stream received at the first subset of the plurality of client devices, wherein the second subset is different from the first subset.
2. The method of claim 1 further comprising: extracting, from the one or more encrypted video streams sent over the network from the content server to the plurality of client devices, a set of stream-related features; extracting, from one or more client devices of the plurality of client devices, a set of quality-related labels corresponding to the quality of experience parameters of the encrypted video stream received at the one or more client devices; and training a classifier using the training dataset to obtain the trained classifier, wherein the training dataset comprises the set of stream-related features and the set of quality-related labels.
This invention relates to monitoring and improving the quality of experience (QoE) for encrypted video streams delivered over a network. The problem addressed is the difficulty in assessing and optimizing video streaming quality when the content is encrypted, preventing direct analysis of the video data. The solution involves extracting stream-related features from encrypted video streams sent from a content server to multiple client devices. These features may include network metrics, packet loss rates, latency, or other indicators that can be derived without decrypting the content. Additionally, quality-related labels are collected from client devices, representing user-perceived quality metrics such as buffering events, resolution changes, or playback interruptions. A classifier is then trained using a dataset that combines these stream-related features and quality-related labels. The trained classifier can predict QoE parameters for future streams, enabling real-time adjustments to improve streaming performance. This approach allows for quality assessment and optimization without compromising encryption, ensuring secure and efficient video delivery.
3. The method of claim 2 , further comprising: validating the trained classifier using the validation dataset; and testing the trained classifier using the test dataset, wherein the validation dataset and the test dataset each comprise a respective set of stream-related features and a respective set of quality-related labels.
This invention relates to machine learning systems for evaluating data streams, specifically focusing on validating and testing a trained classifier to assess its performance. The method involves using a validation dataset and a test dataset, each containing stream-related features and quality-related labels, to evaluate the classifier's accuracy and reliability. The validation dataset is used to fine-tune the classifier, ensuring it generalizes well to unseen data, while the test dataset provides an independent assessment of its performance. The features in these datasets are derived from data streams, and the labels indicate the quality or correctness of the stream data. By systematically validating and testing the classifier with these datasets, the method ensures robust performance in real-world applications where data quality is critical. This approach is particularly useful in domains where data streams must be processed in real-time, such as network monitoring, sensor data analysis, or financial transaction validation. The invention improves upon existing methods by providing a structured way to assess classifier performance, reducing errors and enhancing decision-making based on streamed data.
4. The method of claim 2 , wherein the one or more client devices comprise at least one of controlled user equipment (UEs) or UE simulators.
This invention relates to wireless communication systems, specifically methods for managing and testing user equipment (UE) devices or UE simulators in a network environment. The technology addresses challenges in efficiently controlling and monitoring multiple UEs or UE simulators to evaluate network performance, optimize resource allocation, or validate new protocols. The method involves dynamically configuring and coordinating these devices to simulate real-world usage scenarios, ensuring accurate testing and analysis of network behavior under various conditions. The system may include mechanisms for remote control, data collection, and performance evaluation, enabling operators to assess network reliability, capacity, and user experience. By incorporating both physical UEs and simulated UEs, the approach provides flexibility in testing different network configurations and user behaviors without requiring physical hardware for all test cases. This enhances scalability and reduces testing costs while maintaining high accuracy in network performance assessments. The method supports applications in 5G and beyond, where diverse and complex network conditions must be evaluated to ensure robust service delivery.
5. The method of claim 1 , wherein the at least one stream-related feature comprises at least one of: an effective throughput; a quartile of throughput; at least one of a high, mid, or low peak density; or a total number of the one or more encrypted video streams.
This invention relates to analyzing encrypted video streams in a network to optimize resource allocation and improve performance. The problem addressed is the difficulty in monitoring and managing encrypted video streams without decrypting them, which is often impractical or prohibited due to privacy and security concerns. The solution involves extracting and analyzing stream-related features from the encrypted video streams to infer performance metrics without decrypting the content. The method involves processing one or more encrypted video streams to determine at least one stream-related feature. These features include the effective throughput of the streams, quartiles of throughput (e.g., 25th, 50th, or 75th percentile), peak density metrics categorized as high, mid, or low, and the total number of encrypted video streams being processed. These features are used to assess the performance and quality of the video streams, enabling network operators to optimize bandwidth allocation, detect anomalies, and improve overall streaming efficiency without compromising encryption. By analyzing these non-content-based features, the method provides insights into network performance while maintaining the integrity and security of the encrypted streams. This approach is particularly useful in scenarios where real-time monitoring and adaptive resource management are required, such as in content delivery networks (CDNs), video-on-demand services, and live streaming platforms. The technique ensures that encrypted video streams can be monitored effectively without violating privacy or security protocols.
6. The method of claim 5 , wherein the at least one quality-related label comprises one or more of a rebuffering time percentage, or a streaming reproduction cut-off ratio.
This invention relates to video streaming quality assessment, specifically measuring and labeling streaming performance metrics to evaluate user experience. The method involves analyzing streaming sessions to determine quality-related labels that quantify disruptions in playback. These labels include a rebuffering time percentage, which measures the proportion of total playback time spent buffering, and a streaming reproduction cut-off ratio, which indicates the fraction of the video that was not successfully delivered before playback termination. The method may also involve generating additional quality labels, such as bitrate fluctuations or startup delay, to provide a comprehensive assessment of streaming performance. The labels are derived from real-time or post-session analysis of streaming data, enabling service providers to identify and address quality issues. The invention aims to improve streaming reliability by quantifying and categorizing playback interruptions, allowing for targeted optimizations to enhance user experience. The method can be applied to various streaming protocols and platforms to ensure consistent quality across different delivery systems.
7. A system for determining quality of experience parameters of an encrypted video stream received at a client device, the system comprising: a plurality of client devices configured to receive over a network from a content server, one or more encrypted video streams; a classifier trained using a supervised machine learning algorithm and a training data set, validated based on a validation dataset, and tested based on a test dataset, wherein the classifier is configured to be applied to at least one stream-related feature, extracted from the one or more encrypted video streams without obtaining data directly from the client device and without decrypting the one or more encrypted video streams, to determine at least one quality-related label corresponding to the quality of experience parameters of the encrypted video stream of the one or more encrypted video streams received at the client device, wherein the at least one stream-related feature is extracted from the one or more encrypted video streams via a passive tap into the network, wherein the at least one stream-related feature is associated with at least one of throughput of the one or more encrypted video streams, peak density of the one or more encrypted video streams, or quantity of the one or more encrypted video streams, and wherein the at least one quality-related label is associated with playback quality of the one or more encrypted video streams; and a network operator component configured to: receive the at least one quality-related label from the classifier; change one or more first network parameters associated with the network to perform capacity enhancements at a radio access network used to connect the content server to a first subset of the plurality of client devices via the network or at a network core used to connect the content server to the first subset of the plurality of client devices via the network; analyze the at least one quality-related label determined before and after changing the one or more first network parameters to determine a measurement of an impact of changing the one or more first network parameters on the quality of experience parameters of the encrypted video stream received at the first subset of the plurality of client devices; and change one or more second network parameters associated with the network used to connect the content server to a second subset of the plurality of client devices in view of the measurement of the impact of changing the one or more first network parameters on the quality of experience parameters of the encrypted video stream received at the first subset of the plurality of client devices, wherein the second subset is different from the first subset.
The system monitors and optimizes the quality of experience (QoE) for encrypted video streams delivered to client devices over a network. The system addresses the challenge of assessing video stream quality without decrypting the content or accessing client device data, which is critical for maintaining privacy and security while ensuring high-quality playback. Multiple client devices receive encrypted video streams from a content server. A classifier, trained using supervised machine learning with labeled datasets, analyzes stream-related features extracted passively from the network. These features include throughput, peak density, and the quantity of streams, all obtained without decrypting the video. The classifier assigns quality-related labels indicating playback quality. A network operator component receives these labels, adjusts network parameters (such as capacity enhancements in the radio access network or core network) for a subset of client devices, and measures the impact of these changes on QoE. Based on this analysis, the system further adjusts network parameters for a different subset of client devices to optimize overall performance. This approach enables real-time QoE monitoring and dynamic network optimization for encrypted video streams while preserving privacy and security.
8. The system of claim 7 , wherein the classifier is trained using the training data set comprising a set of stream-related features extracted from the one or more encrypted video streams and a set of quality-related labels received from one or more client devices of the plurality of client devices.
This invention relates to a system for monitoring and classifying the quality of encrypted video streams in real-time. The system addresses the challenge of assessing video quality without decrypting the content, which is critical for maintaining privacy and security while ensuring a high-quality user experience. The system includes a feature extraction module that processes encrypted video streams to extract stream-related features, such as packet loss, latency, and bitrate fluctuations, without decrypting the content. These features are then analyzed by a classifier, which has been trained using a dataset containing similar stream-related features and corresponding quality-related labels provided by client devices. The classifier determines the quality of the video streams based on the extracted features and the training data. The system also includes a feedback mechanism where client devices report perceived quality, which is used to refine the classifier's accuracy over time. This approach enables real-time quality assessment and adaptive adjustments to improve streaming performance without compromising encryption. The invention is particularly useful in secure video streaming applications where privacy and quality assurance are critical.
9. The system of claim 8 , wherein the one or more client devices comprise at least one of controlled user equipment (UEs) or UE simulators.
The invention relates to a system for managing and testing wireless communication devices, specifically user equipment (UE) or UE simulators. The system addresses the challenge of efficiently controlling and monitoring multiple UEs or UE simulators in a network environment, ensuring accurate testing and validation of wireless communication protocols and performance. The system includes a central controller that coordinates the operation of one or more client devices, which may be physical UEs or software-based UE simulators. These client devices interact with a wireless network to perform various tests, such as signal quality assessment, protocol compliance, and network performance evaluation. The central controller manages the configuration, execution, and data collection processes, allowing for automated and scalable testing. The use of UE simulators enables cost-effective and flexible testing scenarios without requiring physical hardware for each test case. This system is particularly useful in research, development, and certification of wireless communication technologies, providing a robust framework for validating network performance and device functionality.
10. The system of claim 8 , wherein the set of stream-related features comprise at least one of: an effective throughput; a quartile of throughput; or at least one of a high, mid, or low peak density; or a total number of the one or more encrypted video streams.
This invention relates to a system for analyzing encrypted video streams, particularly in scenarios where direct access to video content is restricted. The system addresses the challenge of monitoring and managing encrypted video streams without decrypting them, which is critical for applications like network performance optimization, security, and quality of service (QoS) assessment. The system extracts and processes stream-related features from the encrypted video streams to derive insights without compromising security or privacy. The system includes a feature extraction module that identifies key characteristics of the encrypted video streams, such as effective throughput, quartile of throughput, peak density (high, mid, or low), and the total number of streams. Effective throughput measures the actual data transfer rate over time, while quartile analysis provides a statistical breakdown of throughput distribution. Peak density metrics assess the concentration of data packets during high, mid, or low activity periods, helping to detect anomalies or congestion. The total number of streams is tracked to monitor network load and resource allocation. By analyzing these features, the system enables real-time monitoring, predictive maintenance, and adaptive resource management for encrypted video streaming services. This approach ensures efficient network utilization while maintaining compliance with encryption standards. The system is particularly useful in environments where decryption is impractical or prohibited, such as in secure communications or regulated industries.
11. The system of claim 8 , wherein the set of quality-related labels comprises one or more of a rebuffering time percentage, or a streaming reproduction cut-off ratio.
A system for evaluating streaming media quality monitors performance metrics to assess user experience. The system collects data on streaming interruptions, such as buffering events, and calculates a rebuffering time percentage, which represents the proportion of total playback time spent buffering. Additionally, the system determines a streaming reproduction cut-off ratio, indicating the fraction of content that fails to play due to buffering or other disruptions. These metrics are used to generate quality-related labels that quantify streaming reliability and playback continuity. The system may also track other quality indicators, such as bitrate fluctuations or startup delays, to provide a comprehensive assessment of streaming performance. By analyzing these labels, the system enables optimization of streaming protocols, network configurations, or content delivery strategies to improve user satisfaction. The solution addresses the challenge of maintaining seamless playback in varying network conditions, particularly in adaptive bitrate streaming scenarios where buffering and interruptions can degrade the viewing experience. The system's ability to quantify and label streaming quality issues facilitates automated adjustments and targeted improvements in real-time.
12. The system of claim 7 , further comprising a feature extraction module comprising the passive tap that is configured to extract the at least one stream-related feature from the one or more encrypted video streams.
A system for processing encrypted video streams includes a passive tap that monitors and extracts stream-related features without decrypting the content. The system captures encrypted video data from one or more sources and analyzes it to identify characteristics such as packet structure, timing patterns, or metadata without accessing the actual video content. A feature extraction module processes these encrypted streams to derive relevant features, which may include network traffic patterns, packet sizes, or transmission intervals. These extracted features are then used for further analysis, such as monitoring, security, or quality assessment, without requiring decryption. The system operates in real-time, ensuring minimal latency while maintaining the integrity of the encrypted data. This approach enables secure and efficient analysis of video streams in environments where decryption is impractical or prohibited, such as in network monitoring or compliance applications. The passive tap ensures that the original encrypted content remains unaltered, preserving privacy and security while still allowing for valuable insights to be derived from the stream's structural and behavioral attributes.
13. One or more computer-readable storage media having computer-readable instructions stored thereon, which, when executed by a processor, perform operations comprising: extracting, from one or more encrypted video streams sent over a network from a content server to a plurality of client devices, a first instance of at least one stream-related feature associated with at least one of throughput of the one or more encrypted video streams, peak density of the one or more encrypted video streams, or quantity of the one or more encrypted video streams without obtaining data directly from the plurality of client devices and without decrypting the one or more encrypted video streams, wherein the extracting of the first instance of the at least one stream-related feature is via a passive tap into the network; determining a first instance of at least one quality-related label of a plurality of quality-related labels based on applying a trained classifier to the first instance of the at least one stream-related feature extracted from the one or more encrypted video streams, the trained classifier being trained based on a training dataset, validated based on a validation dataset, and tested based on a test dataset, wherein each of the plurality of quality-related labels corresponds to a respective quality of experience parameter of quality of experience parameters of an encrypted video stream received at a client device, and wherein the at least one quality-related label is associated with playback quality of the one or more encrypted video streams; changing one or more first network parameters associated with the network to perform capacity enhancements at a radio access network used to connect the content server to a first subset of the plurality of client devices via the network or at a network core used to connect the content server to the first subset of the plurality of client devices via the network; extracting, from the one or more encrypted video streams sent from the content server to the first subset of the plurality of client devices, a second instance of the at least one stream-related feature after the changing of the one or more first network parameters; determining a second instance of the at least one quality-related label based on applying the trained classifier to the second instance of the at least one stream-related feature extracted after the changing of the one or more first network parameters; analyzing the first instance the at least one quality-related label determined before the changing of the one or more first network parameters and the second instance of the at least one quality-related label determined after the changing of the one or more first network parameters to determine a measurement of an impact of the changing of the one or more first network parameters on the quality of experience parameters of the encrypted video stream received at the first subset of the plurality of client devices; and changing one or more second network parameters associated with the network used to connect the content server to a second subset of the plurality of client devices in view of the measurement of the impact of the changing of the one or more first network parameters on the quality of experience parameters of the encrypted video stream received at the first subset of the plurality of client devices, wherein the second subset is different from the first subset.
This invention relates to monitoring and optimizing video streaming quality in encrypted networks without decrypting the content. The system passively extracts stream-related features such as throughput, peak density, and stream quantity from encrypted video streams transmitted over a network from a content server to multiple client devices. These features are analyzed using a trained classifier to determine quality-related labels corresponding to playback quality and other quality of experience (QoE) parameters. The classifier is trained, validated, and tested using separate datasets. Based on the analysis, network parameters in the radio access network or core network are adjusted to enhance capacity for a subset of client devices. The system then measures the impact of these changes by comparing QoE parameters before and after the adjustments. If improvements are observed, similar network parameter changes are applied to a different subset of client devices to optimize overall streaming performance. The approach avoids direct client device interaction or decryption, enabling real-time QoE monitoring and adaptive network optimization for encrypted video streams.
14. The one or more computer-readable storage media of claim 13 , wherein the operations further comprise: extracting, from the one or more encrypted video streams sent over the network from the content server to the plurality of client devices, a set of stream-related features; extracting, from one or more client devices of the plurality of client devices, a set of quality-related labels corresponding to the quality of experience parameters of the encrypted video stream received at the one or more client devices; and training a classifier using the training dataset to obtain the trained classifier, wherein the training dataset comprises the set of stream-related features and the set of quality-related labels.
This invention relates to monitoring and improving the quality of experience (QoE) for encrypted video streams delivered over a network. The problem addressed is the difficulty in assessing video stream quality when the content is encrypted, making traditional quality measurement techniques ineffective. The solution involves extracting stream-related features directly from encrypted video streams sent from a content server to multiple client devices. Additionally, quality-related labels are collected from the client devices, representing the QoE parameters of the received encrypted streams. These features and labels are used to train a classifier, which can then predict or assess video quality without decrypting the content. The trained classifier leverages the relationship between observable stream characteristics and user-perceived quality, enabling real-time or post-delivery quality analysis. This approach allows content providers and network operators to optimize video delivery performance while maintaining encryption for security. The system improves QoE monitoring in encrypted streaming environments, ensuring high-quality video experiences without compromising data privacy.
15. The one or more computer-readable storage media of claim 14 , wherein the operations further comprise: validating the trained classifier using the validation dataset; and testing the trained classifier using the test dataset, wherein the validation dataset and the test dataset each comprise a respective set of stream-related features and a respective set of quality-related labels.
This invention relates to machine learning systems for evaluating data streams, specifically focusing on validating and testing a trained classifier to assess data quality. The system addresses the challenge of ensuring accurate and reliable classification in streaming data environments, where data quality can vary significantly. The trained classifier is evaluated using separate validation and test datasets, each containing stream-related features and quality-related labels. The validation dataset is used to fine-tune the classifier, while the test dataset provides an independent assessment of its performance. This approach ensures that the classifier generalizes well to unseen data, maintaining high accuracy and reliability in real-world applications. The system may also include preprocessing steps to prepare the data for classification, such as feature extraction and normalization, to enhance the classifier's effectiveness. By systematically validating and testing the classifier, the invention improves the robustness of data quality assessment in streaming environments, making it suitable for applications where data integrity is critical.
16. The one or more computer-readable storage media of claim 14 , wherein the one or more client devices comprise at least one of controlled user equipment (UEs) or UE simulators.
This invention relates to wireless communication systems, specifically to methods for testing and validating user equipment (UE) devices or UE simulators in network environments. The technology addresses challenges in efficiently assessing UE performance, particularly in scenarios where UEs or UE simulators must interact with network infrastructure to verify compliance with communication protocols or standards. The system involves one or more computer-readable storage media storing instructions that, when executed, enable a network testing device to perform operations. These operations include receiving a request to test one or more client devices, which may be controlled UEs or UE simulators. The testing device then configures the client devices to establish a connection with a network, such as a cellular or wireless network, and executes test cases to evaluate the devices' behavior. The test cases may involve simulating network conditions, verifying protocol adherence, or assessing performance metrics like latency, throughput, or signal quality. The results are collected and analyzed to determine whether the client devices meet specified requirements or standards. The invention ensures that UEs or UE simulators can be systematically tested under controlled conditions, improving reliability and interoperability in wireless networks. This approach is particularly useful for manufacturers, network operators, and regulatory bodies to validate device functionality before deployment.
17. The one or more computer-readable storage media of claim 14 , wherein: the set of stream-related features comprises at least one of: an effective throughput; a quartile of throughput; a high, mid and/or low peak density; or a total number of the one or more encrypted video streams; and the set of quality-related labels comprise at least one of: a rebuffering time percentage; or a streaming reproduction cut-off ratio.
This invention relates to monitoring and analyzing encrypted video streams in a network to assess streaming quality and performance. The system extracts and evaluates stream-related features and quality-related labels from encrypted video streams to determine their impact on user experience. The stream-related features include metrics such as effective throughput, quartile throughput, peak density (high, mid, or low), and the total number of encrypted video streams being processed. These features help quantify the network's capacity and efficiency in handling video traffic. The quality-related labels, such as rebuffering time percentage and streaming reproduction cut-off ratio, measure the actual quality of the streaming experience by tracking interruptions and playback failures. By analyzing these metrics, the system can identify performance bottlenecks, optimize network resources, and improve streaming reliability for encrypted video content. The invention is particularly useful in environments where video streams are encrypted, making traditional quality assessment methods ineffective. The solution provides a way to evaluate streaming performance without decrypting the content, ensuring privacy and security while still delivering actionable insights.
18. The method of claim 1 , wherein: the extracting of the first instance of the at least one stream-related feature is without obtaining data directly from the client device and without decrypting the one or more encrypted video streams; and the extracting of the first instance of the at least one stream-related feature is via a passive tap into the network.
This invention relates to a method for extracting stream-related features from encrypted video streams in a network without direct interaction with client devices or decryption of the video content. The method involves passively tapping into the network to monitor and extract these features, enabling analysis of video stream characteristics without compromising security or requiring access to the encrypted data itself. The approach is designed to work in environments where video streams are encrypted end-to-end, such as in secure communication systems or content delivery networks, where traditional monitoring techniques would require decryption or client-side data access. By using a passive network tap, the method avoids altering network traffic or requiring cooperation from client devices, making it suitable for real-time or forensic analysis of video streams. The extracted features may include metadata, packet patterns, or other indicators that can be used for quality assessment, security monitoring, or network optimization without exposing the actual video content. This technique is particularly useful in scenarios where privacy and security are critical, as it allows for network-level analysis without decrypting or accessing the encrypted payloads.
19. The method of claim 18 , wherein: the extracting of the second instance of the at least one stream-related feature is without obtaining data directly from the client device and without decrypting the one or more encrypted video streams; and the extracting of the second instance of the at least one stream-related feature is via the passive tap into the network.
This invention relates to a method for extracting stream-related features from encrypted video streams in a network without direct interaction with client devices or decryption of the video content. The method involves passively monitoring network traffic to identify and extract specific features from encrypted video streams, such as packet headers, metadata, or other non-payload data, without accessing the encrypted payload itself. This approach allows for analysis of video stream characteristics, such as quality, performance, or usage patterns, while maintaining privacy and security by avoiding decryption or direct client device interaction. The passive network tap enables real-time or near-real-time extraction of these features, facilitating monitoring, optimization, or troubleshooting of video streaming services without compromising the confidentiality of the encrypted content. The method is particularly useful in scenarios where encrypted video streams are transmitted over networks, and there is a need to analyze stream-related data without decrypting the content or accessing client devices. By leveraging passive network monitoring, the invention provides a non-intrusive way to gather stream-related insights while adhering to privacy and security constraints.
20. The method of claim 1 , wherein the measurement of the impact is a metric associated with a difference between the first instance of the at least one quality-related label and the second instance of the at least one quality-related label.
This invention relates to assessing the impact of changes in data quality within a system. The problem addressed is the need to quantify how modifications to data attributes affect overall data quality, particularly in systems where labeled data is used for training machine learning models or other analytical processes. The invention provides a method to measure the impact of changes by comparing two instances of quality-related labels associated with the same data. The first instance represents the original state of the data, while the second instance represents the modified state after an intervention or update. The impact is determined by calculating a metric that quantifies the difference between these two instances. This metric can be used to evaluate the effectiveness of data quality improvements, identify unintended consequences of changes, or validate the reliability of updated data. The method ensures that data quality assessments are objective and measurable, supporting better decision-making in data-driven applications. The invention is particularly useful in fields like machine learning, data governance, and quality assurance, where maintaining high-quality data is critical for accurate analysis and model performance.
Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.
July 31, 2018
January 25, 2022
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.